Note: Descriptions are shown in the official language in which they were submitted.
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
METHOD FOR REGULATING SIZE AND GROWTH
OF VASCULARIZED NORMAL TISSUE
Field of the Invention
The present invention is directed generally to the field of treatment of
obesity and other disorders characterized by proliferation of normal
vascularized tissues, by the administration of effective amount of
angiogenesis inhibitors.
Background of the Invention
The prevalence of overweight has reached epidemic proportions in
most developed countries and carries with it staggering mortality and
morbidity statistics. Obesity is a well established risk factor for a number
of
potentially life-threatening diseases such as atherosclerosis, hypertension,
diabetes, stroke, pulmonary embolism, and cancer. (Meisler J., St. Jeor S.
1996. Am J Clin Nutr. 63:409S-411 S). (Bray G. 1996. Endocrin Metab Clin
North Amer. 25:907-919). Furthermore, it complicates numerous chronic
conditions such as respiratory diseases, osteoarthritis, osteoporosis, gall
bladder disease, and dyslipidemias. The enormity of this problem is best
reflected in the fact that death rates escalate with increasing body weight.
More than 50% of all-cause mortality is attributable to obesity-related
conditions once the body mass index (BMI) exceeds 30 kg/m2, as seen in 35
million Americans. (Lee L, Paffenbarger R. 1992. JAMA. 268:2045-2049).
By contributing to greater than 300,000 deaths per year, obesity ranks second
only to tobacco smoking as the most common cause of potentially
preventable death. (McGinnis J., Foege W. 1993. MA.270:2207-2212).
Accompanying the devastating medical consequences of this problem
is the severe financial burden placed on the health care system in the United
States. The estimated economic impact of obesity and its associated illnesses
from medical expenses and loss of income are reported to be in excess of $68
billion/year. (Colditz G. 1992. Am J Clin Nutr. 55:503S-507S). (Wolf A.,
Colditz G. 1996. Am J. Clin Nutr. 63:466S-469S). (Wolf A., Colditz G.
1994. Pharmacoeconomics. 5:34-37). This does not include the greater than
$30 billion per year spent on weight loss foods, products, and programs.
(Wolf A., Colditz G. 1994. Pharmacoeconomics. 5:34-37). (Ezzati, et
a.1992. Vital health Stat [2]. 113).
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
In 1990, the US government responded to the crisis by establishing as
a major national health goal the reduction in the prevalence of obesity to
(20% of the population by the year 2000. (Public Health Service. Healthy
people 2000: national health promotion and disease prevention objectives.
1990. (US Department of Health and Human Services Publication PHS 90-
50212.))
In spite of this objective, the prevalence of overweight in the United States
has steadily increased, reaching an astounding 33.0% in the most i-ecent
National Health and Nutrition Examination Survey (1988-1991).
(Kuczmarski, et al.. 1994. JAMA. 272:205-211). Furthermore, the mean
BMI has also increased over this period by 0.9 kg/m2. This alarming trend
has not occurred as the result of lack of effort. On the contrary, an
estimated
25% of men, 50% of women, and 44% of adolescents are trying to lose
weight at any given time. (Robinson, et al.J Amer Diabetic Assoc. 93:445-
449). Rather, the 31 % increase in rate and 8% increase in prevalence over
the past decade is a testimony of the fact that obesity is notoriously
resistant
to current interventions. (NIH Technology Assessment Conference Panel.
1993. Ann Intern Med. 119:764-770).
A major reason for the long-term failure of established approaches is
their basis on misconceptions and a poor understanding of the niechanisms
of obesity. Conventional wisdom maintained that obesity is a self-inflicted
disease of gluttony. Comprehensive treatment programs, therefore, focused
on behavior modifications to reduce caloric intake and increase physical
activity using a myriad of systems. These methods have limited efficacy and
are associated with recidivism rates exceeding 95%.
Failure of short-term approaches, together with the recent progress
made in elucidating the pathophysiology of obesity, have lead to a
reappraisal of pharmacotherapy as a potential long-term, adjuvant treatment.
(National Task Force on Obesity. 1996. JAMA. 276:1907-1915). (Ryan, D.
1996. Endo Metab Clin N Amer. 25:989-1004). The premise is that body
weight is a physiologically controlled parameter similar to blood pressure
and obesity is a chronic disease similar to hypertension. The goal of long-
term (perhaps life long) medical therapy would be to facilitate both weight
loss and subsequent weight maintenance in conjunction with a health.N diet
2
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
and exercise. To assess this approach, the long-term efficacy of cu--rently
available drugs must be judged against that of non-pharmacological
interventions alone. The latter approach yields an average weight loss of 8.5
Kg at 21 weeks of treatment and only maintains 50% of the weight reduction
at 4 years in 10-30% of the patients. (Wadden T. 1993. Ann Intern Med.
119:688-693). (Kramer, et al.1989. Int J Obes. 13:123-136). The few
studies that have evaluated long-term (greater than 6 months) single-drug
(Guy-Gran, et al.. 1989. Lancet. 2:1142-1144) (Goldstein, et al. 1994 Int J
Obes. 18:129-135) (Goldstein, et al. 1993. Obes Res. 2:92-98) or
combination therapy (Weintraub M. 1992. Clin Pharmacol. Ther. 51:581-
585) show modest efficacy compared with placebo in the reduction of body
weight.
All medications currently used to treat or prevent obesity ~urc di,-ected
at the adipocyte compartment of the tissue and work by either decreasing
energy availability or increasing energy output. These agents can be placed
into three categories based on mechanism. (National Task Force on Obesity.
1996. JAMA. 276:1907-1915).
Reduction of energy intake. This approach is directed at reducing
food intake by decreasing appetite or increasing satiety. These 'anorexiant'
drugs affect neurotransmitter activity by acting on either the
catecholaminergic system (amphetamines, benzphetamine, phendimetrazine,
phentermine, mazindol, diethylpropion, and phenylpropanolamine) or the
serotonergic system (fenfluramine, dexfenfluramine, fluoxetine, sertraline,
and other antidepressant selective serotonin reuptake inhibitors [SSRI]).
Reduction in absorption of nutrients: Drugs in this category block
the action of digestive enzymes or absorption of nutrients. An example of
this type of drug is orlistat which inhibits gastric and pancreatic iipase
activity. (Drent M., van der Veen E. 1995. Obes Res. 3(suppl 4):623S-
625S). These medications are experimental in the United States and not
available for the treatment of obesity.
Increase in enerjzy expenditure: An increase in energy expenditure
may be accomplished by increasing metabolic rate, for example, through
changes in sympathetic nervous system tone or uncoupling of oxidative
phosphorylation. Drugs that affect thermogenesis-metabolism include
3
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
ephedrine alone and in combination with caffeine and/or aspirin, (Passquali
R., Casimirri F. 1993 Int J Obes. 17(suppl 1):S65-S68) and BRL 26830A, a
(-adrenoceptor agonist. (Connacher, et a1.1992. Am J Clin Nutr. 55:258S-
261 S). This class of medications is not approved by the FDA for weight
control.
Currently, no single drug regimen emerges as superior in either
promoting or sustaining weight loss.
Surgical interventions, such as gastric partitioning procedures,
jejunoileal bypass, and vagotomy, have also been developed to treat severe
obesity. (Greenway F. 1996. Endo Metab Clin N Amer. 25:1005-1027).
Although advantageous in the long run, the acute risk benefit ratio has
reserved these invasive procedures for morbidly obese patients according to
the NIH consensus conference on obesity surgery (BMI greater than 40
kg/m2). (NIH Conference. 1991. Ann Intern Med. l 15:956-961). Therefore,
this is not an alternative for the majority of overweight patients unless and
until they become profoundly obese and are suffering the attendant
complications.
There is no medical or surgical treatment for obesity that is directed
at the vascular compartment of the tissue.
It is therefore an object of the present invention to provide a
treatment to reduce obesity.
It is a further object of the present invention to provide compositions
for treatment of obesity.
It is another object of the present invention to provide methods for
inhibiting or decreasing overproliferation of other normal, vascularized
tissues.
Summary of the Invention
Angiogenesis inhibitors are administered to patients in an aniount
effective to regulate normal, non-transformed vascularized tissue size and/or
growth by regulating its vascular compartment. Examples of tissues that can
be controlled include adipose tissue, intestinal polyps, muscle (including
cardiac) tissue, and endometrial tissue. The response of these tissues to the
angiogenesis inhibitors is dose-dependent, reversible, and common to a
variety of different angiogenesis inhibitors (examples use TNP-470,
4
CA 02307792 2005-06-08
angiostatin, and endostatin), based on studies in animal models of obesity,
intestinal polyps, cardiac hypertrophy, and endometriosis. Initial studies
conducted in an adipose tissue model (genetically obese mice and normal
control mice) sliowed that the growth and mass of adipose tissue is under the
control of microvascular endothelium. Expansion of adipose tissue was
associated witli endothelial cell proliferation. Inhibition of angiogenesis
led
to reduction in adipose tissue mass. Weight gain in animals receiving
angiogenesis inhibitors was significantly restricted, in spite of increases in
appetite sufficient to cause weight gain in paired-fed mice. Discontinuation
of the inhibitor resulted in rapid expasion of the adipose tissue. The effect
was dose-dependent, repeatedly reversible, and occurred in response to all of
the inhibitors tested. Significant inhibition was also observed in both the
intestinal polyp and cardiac hypertrophy animal models, using dosages of
two-thirds or less than the dosages used to treat tumors. Preliminary results
in an endometriosis model also show a clear trend towards decreased
development of endometriosis in animals treated witll angiogenesis inhibitors
at a dosage of one-third the dosage used to treat tumors. No effect on normal
tissue that was not proliferating, other than adipose tissue, was observed.
The angiogenesis inhibitor may be a collagenase inhibitor.
Brief Description of the Drawing
Figure IA is a graph of body weight over time (days) in obiOb mice
treated with TNP-470 at either 5 mb/kg daily or 10 m~/kg dailv compared to
ob/ob mice treated with fenfluramine [40 mg/kg intraperitoneally daily] and
saline controls.
Figure 1B is a graph of body weight over time (days) in ob/ob mice
treated with TNP-470 (10 mg/kg daily) paired feeding [explain], and saline
controls.
Figure 2 is a graph of body weight (grams) versus days of treatment,
of mice at doses of 2.5, 5.0, 7.5 and 10 mg TNP-470 mg/kg/day versus
control. n= 4 each group other than 10 mg, which is n = 8.
Figure 3 is a graph of body weight (grams) versus days of treatment,
of mice, with angiostatin (20 mg/kg/day), endostatin (50 mg/kg/12 h), and
angiostatin (50 mg/kg/12 h) as compared to control treated with saline. n=3
each group.
Figure 4A is a graph of body weight 9grams) versus days oE'
5
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
treatment for ob/ob mice treated with 10 mg TNP-470/kg/day (n=6), versus
control ob/ob treated with saline (n=4), and normal mice 9C57B 1/6 treated
with saline (n=5).
Figure 4B is a graph of body weight (grams) versus days of treatment
for ob/ob mice treated with 10 mg TNP-470/kg/day compared with control
saline treatment. n=6 for each group.
Figure 4C is a graph of body weight (grams) versus days, for ob/ob
animals treated with TNP-470 at a dose of 10 mg/kg/day (n=15), versus
normal C57/B1/6 animals treated with saline (n=5) and ob/ob animals treated
with saline (n=4).
Figure 5 is a graph of body weight (grams) versus days of treatment
of ob/ob animals treated with fenfluramine (40 mg/kg/day, n=4) alone and
subsequently with TNP-470 (10 mg/kg/day, n=4).
Figures 6A and 6B are graphs of body weight (grams) versus day of
treatment with TNP-470 (10 mg/kg/day) versus saline controls, in paired
feeding experiments.
Figure 6C is a graph of cell number/well (x 103) versus TNP-470
(pg/ml) when growth factors, BCE and FGF or 3T3 and FGF, were added to
the cells.
Figure 7 is a graph of ventricular weight as %body weight of animals
with thyronine induced ventricular hypertrophy, treated with T3 (induces
hypertrophy), T3 and TNP-470 at 10 mg/kg qod, and T3 and TNP-470 at 20
mg/kg qod.
Figure 8A is a graph of the number of polyps in Min/+ mice, saline
controls versus animals treated with TNP-470 at a dose of 10 mg/kg qod,
after one and three weeks.
Figure 8B is a graph of the size distribution of polyp size in Min/+
mice, graphed as total number of polyps of each size (diameter in mm) for
animals treated with TNP-470 at 10 mg/kg qod, versus saline controls.
Figure 9 is a graph of tissue diameter (mm) in an animal model of
endometriosis for control animals and animals treated with TNP-470.
Detailed Description of the Invention
Angiogenesis is the process by which new blood vessels are formed.
It is essential in reproduction, development, and wound repair, and is highly
6
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
regulated in these normal events. However, unregulated aiid persistent
angiogenesis also contributes to many disease states such as arthritis, ocular
neovascularization, and solid tumor growth. Since 1983, several molecules
have been identified that either stimulate or inhibit angiogenesis. These
molecules have lead to clinical applications and trials in the areas of
diagnostic studies, acceleration of angiogenesis in wound healing, and
inhibition of angiogenesis in neoplasia.
Angiogenesis has been extensively studied in the area of solid tumor
growth and many of the regulatory factors have been isolated from these
tissues. There is substantial evidence that the size and growth of solid
tumors is angiogenesis dependent. Simply stated, once a tumor has been
initiated, every increase in the tumor cell population must be proceeded by
an increase in new capillary formation to supply that tumor This association
has never been made between angiogenesis and 'normal' tissues or oi-gans,
such as adipose tissue.
Adipose tissue is a highly vascularized tissue and has been exploited
to facilitate wound healing for centuries, (Ball J. 1928. The sack-'em-up
men: an account of the rise and fall of the modern resurrectionists. Oliver
and Boyd, Edinburgh, London. p. 73) primarily due to its ability to promote
revascularization of the affected tissue. As a result of this feature, adipose
tissue continues to be utilized in surgical procedures involving a variety of
tissues in which a vascular supply'is vital, such as the intestines,
myocardium, and transplanted lung. (Silverman, et al.. 1988. Angiogenic
activity of adipose tissue. Bioch Biophy Res Com. 153:347-352).
Investigations into the angiogenic aspects of adipose tissue revealed a
growth dependence upon the development of a vascular supply. During
embryogenesis, vascularization and adipocyte differentiation are tightly
coordinated events. Newly formed adipose tissue requires continued
angiogenesis for further growth and development. (Wasserman F. 1965.
The development of adipose tissue. In: Handbook of Physiology. Vol 5.
Renold A., Cahill G., (eds.). Washington, DC. Am Physiol Soc. PP. 87-
100). Although the mechanisms involved are not yet fully elucidated,
bidirectional paracrine interactions between adipocytes and microvascular
endothelial cells (EC) are known to play a key regulatory role. (Lau D.,et al.
7
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
1996. Intern J Obes. 20 (Supp13) S16-S25). This is supported by in vitro
studies evaluating the interactions of EC isolated from adipose tissue with
3T3 cells, an adipocyte line that maintains most of the attributes of fat
cells
in vivo. (Green H. 1978. The adipose conversion of 3T3 cells. In: 10th
Miami Symposium on Differentiation and Development. Ahead F., Schultz
J., Russe T., Wagner R. (ed.). New York, Academic Press. pp 13-33))
These studies demonstrate that both EC and preadipocytes release mitogenic
factors that enhance the proliferation of each other in a cell-specific
manner.
This symbiotic relationship can be seen in co-cultures, where preadipocytes
proliferate in'islets' around groups of EC. (Lau, et al.. 1996. Intern J Obes.
(Supp13) S16-S25).
Progress has been made in identifying many of the compounds
involved. 3T3 adipocytes produce potent factors that stimulate EC
proliferation and migration in vitro and angiogenesis in the chick
15 choroiallantoic membrane. (Cystolith, et al.. 1982. Proc Natl Acad Sci USA.
77:6007-6011). (Cystolith, et al. 1982. Proc Natl Acad Sci USA. 7:5597-
5601). One such compound has been identified as 1-butyryl-glycerol, an
angiogenic factor whose levels increase at least 200-fold during adipocyte
differentiation. (Dobson, et al. 1990. Cell. 61:223-230). Adipocytes also
20 produce other factors with known angiogenic activity, including.: PGE I.
PGE2, TNF-(, bFGF, and FGF-9. (Form D., Auerbach R. 1983. Proc Soc
Exp Biol Med. 172:214-218). (Smith S. 1996. Endo Metab Clin N Amer.
25:921-942). (Teichert-Kuliszewska, et al. 1992. J Clin Invest. 90:1225=
123 1). (Samagh, et al.1995. Obes Res. 3:370S). Reciprocally, EC isolated
from adipose tissue release mitogens into the culture media that stimulate
preadipocyte replication in a dose-dependent fashion. (La, et al. 1996. Intern
J Obes. 20 (Suppl 3) S 16-S25). Partial purification of the EC derived
mitogens have revealed several active proteins with molecular wei"hts
ranging from 18-67 kDa, some of which may belong to the heparin-binding
3o FGF family. (Lau, et al.1990. Int J Obes. 14:193-201). The complex
autocrine/paracrine communications between adipose tissue and EC remains
an active area of investigation.
The studies described in the examples provide irrefutable evidence
that adipogenesis is dependent on the growth of new blood vessels. By
8
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
extension and analogous to solid tumor growth - prevention of
neovascularization using an anti-angiogenic agents will in turn prevent the
growth of adipose tissue. Surprisingly, however, it is also apparent from the
examples that angiogenesis can cause weight loss in normal animals, not just
in animals in which there is an increased proliferation of adipose tissue.
These results provide strong evidence that administration of angiogenesis can
be used to regulate size as well as growth of adipose tissue.
Additional studies described in the animals provide further evidence
that administration of angiogenesis can be used to regulate the size and/or
lo growth of normal vascularized tissues, without harm to the tissues As used
herein, "normal" tissues refers to tissues which are not cancerous, i.e.,
transformed, or scar tissue arising from wound healing. As used herein,
"vascularized tissue" refers to tissues such as muscle, intestine, adipose
tissue, and endometrium, which are normally characterized by a blood
supply providing for nutrient and gas exchange throughout the tissue.
"Vasculature", in contrast to vascularized tissue, refers to the blood vessels
themselves.
1. Compositions for Decreasing Adipose Tissue
A. Compounds Inhibiting Angiogenesis mechanisms
Any anti-angiogenic compound can be used to regulate the size
and/or growth of normal vascularized tissue. Preferred exemplary anti-
angiogenic compounds include TNP-470, described by U. S. patent No.
5,290,807; Angiostatin, described by U.S. Patent No. 5,639,725; Endostatin,
and Thalidomide. Other compounds are known to those skilled in the art.
The following list of angiogenic inhibitors was published in Genetic
Engineering News, October 1, 1998:
9
O
~o
TABLE 1 EXAMPLES OF ANGIOGENIC INHIBITORS AND STIMULATORS IN DEVELOPMENT
Develo ment Stage
Product C1ass Product Name Company Product Description Clinical Indications N
Angiogenic avP3 Antagonist SmithKline Beecham Orally active av03 integrain
Cancer, osteoporosis, Preclinical
Inhi.bitors antagonist R.A. restenosis, ocular
disease
Angiostatin EntreMedJBristol- 38 kD fragment of Cancer Preclinical
Meyers Squib IasminoQen
Angiostatin- Genetix Retroviral delivery of DNA Cancer Preclinical
endostatin lusion Phannaceuticals encoding angiostatin-
rotein endostatin fusion protein Anti-Flk/KDR ImClone Svstems MAb against Flk-
1/KDR Cancer Preclinical
Anti-VE- IinClone Systems MAb against endothelial cell Cancer Preclinical
Cadherins junction VE cadherins;
prevents assembly of
endothelial cells into viable
blood vessels
Combretastotin A4 OxiGene Natural compound derived Solid tutnors Preclinical
1 from bark of a South African
tree; disrupts tumor 4~,
vasctilature by binding to
tubuitt
Endostatin EntreMed 20 kD fragment of collagen Cancer Preclinical
XVIII
Flt-1 Merck Research Lab. Soluble Flt-1 VEGF receptor Cancer Preclinical
K5 Abbott K5 fragment of human Cancer Preclinical
plasminogen 2-niethoxy- EntreMed biducer of endothelial cell Breast cancer
Preclinical
estradiol a o tosis
PD-173074 Parke-Davis bFGF receptor tyrosine kinase Cancer Preclinical
inlubitor .ti
RG8803 Repligen Corp. Heparin-like carbohydrate: Solid tumc .,uif i
Preclinical
inlubits the binding of VEGF diabetic retinopailn
and hFGF to endothelial cell- rn
s ecitic carbohydrate receptors oo
S-836, SC-68448 Monsanto USA av 3 inte n anta onists Solid tumors Preclinical
00
0
Product Class Product Name Com anv Product Description Clinical Indications
Develo ment Stage TBC-1635, TBC- Texas Biotechnology Small molecule antagonist
of Diabeti.c retinopathv, Preclinical io
2653, TBC-3685 Corp. VEGF cancer, psoriasis,
Small molecule antagonist of arthrifis
Inte rin a4 1
Troponin I Boston Life Sciences Recombinant human Troponin Cancer and ocular
Preclinical
1; thought to act as growth diseases
factor rece or antagonist
ANGIOZI'MERibozyme Catalytic RNA that Cancer IND filed
Phannaceuticals specifically cleaves Flt-1
mRNA >
NX 1838 NeXstar Oligonucleotide (aptamer) Age-related macular IND filed
7/22/98
Phannaceuticals antagonist of VEGF degeneration Neoretna Aeterna Laboratories
AE-94 1, a water-soluble Macular degeneration Phase I(Canada) -4
4
Psovascar extract of shark cartilage; Psoriasis Phase I (Canada) Neovastat
inhibits activity of matrix Refractory cancers of Phase I (U.S. &
metalloproteinase II lung; prostate and Canada)
breast
I MI)-121974 Merck KGaA Cyclic peptide antagonist of Cancer Phase I(Germanv)
the V<35 integrin 4~,
Metastat CollaGenex Non-antimicrobial analo_ue of Metastatic cancers Phase I
Pharmaceuticals tetracycline; inhibits activity -4
of matrix metallo roteinase II
Pl-RR Progen Industries, Ltd. Sulfated maimopentasose; acts Solid tumors,
Phase I (U.K.) for
as a heparan mimetic, prevents prevention ol solid tumors
release of activated fonns of restenosis, treatnient Preclinical for other
VEGF and bFGF, and, of asthma and indications
therefore, inhibits angio- intlammatorv ho"el
genesis. Binds to heparanase, disease
blocks breakdown of vascular
menibrane <+nd inhibits _ metast~isis A
Syunlauninc Magainin Slrirk cartilage-derived Solid tuincirs Pliase I wq
I'hannaceuticals inhibitor of;~rowth liictor- Cn
mediated endothelial cell o"Oo
-- - - roliferatioit
00
C
Product Class Product Name Company Product Description Clinical Indications
Development Stage zz
Vitaxin Ixsys MAb against cxv03 integrin; Cancer Phase I
leads to p53-dependent
a o tosis of endothelial cells
SU5416 Sugen Inc. Small molecule inhibitor of Solid tumors, Phase UII
Flk-1 JKDR including KS
Anti-VEGF Genentech MAb against VEGF Solid tumors Phase 11
Thalidomide EntreMed/Bristol- Small molecule inhibitor of Brain cancer, KS,
Phase II
Meyers Squibb grov,4h factor-niediated metasta6c prostate
endothelial cell preliferation cancer and macular
degeneration >
TNP470 TAP Holdings, Inc. Fuinagillin analooue Cancer, arthritis, Phase lI
psoriasis, aud ocular
disease
ZD()101 Zeneca Polysaccharide endotoxin; Solid tumors Phase II
Phacmaceuticals stimulates an inunune-
mediated anti-amgiogenic
N response Neovastat Aetema Laboratones (see above) Solid tumors Phase IUIII
(Canada)
AG334(1 Agouron Matrix metalloproteinase Non-SCLC and Phase }II 4~,
Phannaceuticals iffliibitor hormone-refractory N
prostate cancer
MarimastatTM British Biotech Inc. Matrix inetalloproteinase Ovarian, small
cell Phase III
iiilhibitor and non-small cell
lung, pancreatic and
gastic cancers and
glioblastoma
Another angiogenesis inhibitor that can be used is CM 101, a hacterial toxin
product, being developed by CarboMedi, Inc Brentwood,
TN
ro
z
~o
00
~
N
~O
~O
00
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
B. Carriers/routes/means for administration:
Drugs can be administered parenterally or enterally. In the preferred
embodiment, drugs are administered orally, in an enteric carrier if necessary
to
protect the drug during passage through the stomach. Alternative methods of
delivery include intravenous, transbuccal or other trans-membrane delivery,
and
controlled release formulations.
C. Dosages:
The anti-angiogenic composition is administered in an amount effective
to regulate the size and/or growth of a vascularized tissue. The effective
amount
will be typically an amount effective to limit tissue proliferation or to
decrease
tissue size, especially in the case where the tissue is adipose tissue
Compositions as used herein contain an effective amount of angiogenesis
inhibitor to treat a patient to achieve the desired regulation in the
substantial
absence of systemic toxicity.
U. Methods of Treatment
A. Proposed treatment schedules
As demonstrated by the examples, the angiogenesis inhibitor is
administered in an amount and time period which results in blood levels
regulating the size and/or growth of the vascularized tissue to be treated. ln
the
preferred embodiment for the treatment of obesity, patients will receive drug
once daily in a dosage effective to decrease the weight to maintenance levels.
B. Types of patients
The method of treatment has been demonstrated to be applicable to both
normal overweight individuals and individuals with genetic defects. The
method should also be useful in most cases involving weight gains due to
hormonal or metabolic defects or drug side effects. In addition to promoting
loss of body fat while maintaining lean body mass and being able to sustain
weight loss during chronic administration, other benefits of the treatment
include normalization of blood glucose levels in obesity related diabetes, and
in
the case of some drugs such as TNP-470, reduction of appetite (i.e., as an
anorexic agent). These results can be obtained without unwanted side-effects,
13
CA 02307792 2000-04-27
WO 99/39702 PCTIUS98/22998
with no potential for abuse or dependence.
Angiogenesis dependence of normal organ size and growth has
implications in the areas where excess or insufficient tissue growth results
in
clinical sequela, as well as in the field of tissue engineering. The exaniples
demonstrate efficacy in animal models of endometriosis, polyposis, cardiac
hypertrophyy, as well as obesity. These results are indicative of efficacy for
the
treatment of disorders involving increases in size or growth of other types of
normal tissue, such as uterine fibroids, prostatic hypertrophy, and
amyloidosis.
These disorders typically involve patients where increases in size or growth
of
lo tissue is not normal, but the tissue is not transformed or of a completely
different
type of tissue than the tissue of origin, although the location (as in thc
case o t'
endometriosis) or growth (organ hypertrophy) is abnormal.
III. Examples
The present invention will be further understood by reference to the
following non-limiting examples.
Example 1: Treatment of genetically obese mice with TNP-470
The first series of experiments were performed on ob/ob mice, a strain of
mice that develops morbid obesity as a result of a spontaneous mutation in the
leptin gene. (Zhang, et al. 1994. Nature. 372:425-432). Animal studies were
carried out in the animal facility of Children's Hospital of Boston in
accordance
with the institutional guidelines. Male ob/ob mice (Jackson Laboratories, Bar
Harbor, ME) with ages ranging from 7 weeks to 6 months and weights ranging
from 35 - 70 grams were acclimated before experimentation, individually caged,
and had free access to water throughout the studies. Animals were anesthetized
with inhaled methoxyfluorane (Pitman-Moore, inc., Mundelein, IL) prior to all
procedures (shaving, and photography) and were monitored until fuilv awake.
The animals were sacrificed at the end of the experiments by continuous
inhalation of methoxyfluorane.
The backs of the mice were shaved to facilitate injections. All injections
and measurements were performed between 4-6 p.m. each day. The mice were
injected subcutaneously with 5 or 10 mg/kg of TNP-470 (AGM-1470) or saline
14
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
control using a 30G needle for treatment periods up to 30 days (thus far). The
subcutaneous fat was observed to decrease at the site of the TNP-470
injections
so the site was varied along the dorsum. For comparison, one group of animals
received fenfluramine (40 (:g/kg, intraperitoneally, every day). The wei~ht of
each animal and the food remaining were measured each day. One group of
animals was paired fed to determine if all or part of the weight loss effect
of
TNP-470 was the result of appetite suppression. This was done by pairing an
untreated mouse with each treated mouse and feeding the untreated mouse the
amount of food its treated partner ate the day before.
Figure lA demonstrates the weight loss response of mice treated with
TNP-470. The treated mice lost weight in a dose dependent fashion. The
weight loss in the TNP-470 treated groups was significantly greater than that
achieved by fenfluramine at maximal dose. The weight loss of the group
receiving 10 mg/kg of TNP continued at an average of 1 gram per day until they
reached a lean body mass of approximately 22-24 grams. This weight loss was
achievable independent of the weight of the animal at the start of the study.
The
weights of the animals stabilized at these levels in spite of continued drug
administration. The animals receiving a lower dose of TNP-470 stabilized at a
higher weight. The animals remained healthy and active throughout the study.
As shown by Figure 1B, mice treated with TNP-470 had a decrease in
appetite compared with control animals that alone resulted in loss of weight.
However, the weight loss resulting from the decrease in food intake was
significantly less than that achieved by TNP-470. These results suggest that
TNP-470 is acting via two mechanisms to reduce weight - a central appetite
suppression and a peripheral angiogenesis inhibition. The appetite suppression
effect of TNP-470 has never before been reported.
Example 2: Treatment of normal mice to decrease fat by
administration of TNP-470.
The paired feeding experiment was repeated using normal rnice without
the leptin gene mutation that did not develop obesity. In these animals, which
began the study relatively closer to their lean body mass, a small wei(yht
loss
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
was observed that was similar in the treated and the paired fed groups. This
suggests that in a lean animal, in which the angiogenesis occurring in the
slowly
growing adipose tissue is expected to be limited, the majority of the weight
loss
achieved with TNP-470 is the result of appetite suppression. These animals
also
showed no evidence of toxicity.
Samples of adipose tissue were examined histologically for evidence of
endothelial cell proliferation indicating new blood vessel formation in both
treated and untreated animals using PCNA and Brdu staining methods. There
was significantly greater endothelial cell proliferation in the untreated
animals
compared with those receiving TNP-470. This supports the hypothesis that
there is angiogenesis in the adipose tissue of these mice and that it can be
inhibited by TNP-470.
In conclusion, TNP-470 causes a dose dependent loss of weight in both
obese and normal mice to the level of lean body weight as the result of both
an
appetite suppression and inhibition of angiogenesis independent of the
starting
weight of the animal. The amount of weight loss achieved with TNP-470 is far
superior to that achieved with fenfluramine. These results support the use of
angiogenesis inhibition for the treatment of obesity. ln addition, an appetite
suppression effect of TNP-470 is alsoestablished.
Experimental Procedures
Animal Studies
Animal studies were carried out at Children's Hospital, Boston in
accordance with institutional guidelines. Male ob/ob (C57BL/6J) and C57
(C57BL/6J) mice were purchased at 7 weeks of age and acclimated for one
week before experimentation (Jackson Laboratories, Bar Harbor, ME). Mice
were housed individually with free access to water and standard chow.
Manipulations were performed between 4-6p.m. to minimize disruption of the
feeding routine. Animals were anesthetized with inhaled methoxyfluorane
(Pitman-Moore, Inc., Mundelein, IL) prior to procedures causing discomfort or
distress and were monitored until fully awake. The hair was shaved to
facilitate
injections. TNP-470 was generously provided by Takeda Chemical I ndustries.
16
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
Ltd. (Osaka, Japan) and stored dry at 40 C. Solutions were prepared daily in
normal saline and administered subcutaneously in rotating sites along the
dorsum. Body weight and chow remaining in the feeders were measured daily.
Animals were sacrificed by continuous inhalation of CO2.
Dose Response Sturly Eight week old, ob/ob mice were treated with 2.5, 5.0,
7.5, or 10 mg/kg/day of TNP-470 for 21 days. A control group received 0.1
ml/day of saline. The group that received 10 mg/kg/day of TNP-470 contained
8 mice, all other groups contained 4 mice each.
Treatment of Morbidly Obese Mice Eight week old ob/ob mice were housed
until they completed their rapid growth phase and weighed greater than 60 gm.
At approximately 5 months of age (n=6), treatment was initiated with I U
mg/kg/day of TNP-470 and continued for 67 days. An age-matched control
group of ob/ob mice (n=6) received 0.1 ml/day of saline.
Long-terni Treatment Sturly Eight week old (n=6), ob/ob/ mice were treated
with 10 mg/kg/day of TNP-470 for 138 days. A control group (n=4) received
0.1 ml/day of saline. Body weights of normal, age-matched C57 mice (n=5)
were recorded daily for comparison.
Comparison with Fenfluramine Eight week old, ob/ob mice were injected
intraperitoneally with 40 mg/kg/day of fenfluramine in 0.1 ml of saline for 45
days (n=12). The mice were then evenly divided into 3 groups and were either
1) taken off therapy, 2) continued on fenfluramine, or 3) changed to 10
mg/kg/day of TNP-470 for the remainder of the 168 day study. A control group
of ob/ob mice (n=4) received 0.1 ml/day of saline.
Paired Feeding Studies Eight week old, ob/ob mice with free access to chow
were treated with 10 mg/kg/day of TNP-470 (n=4). The chow remaining in the
feeder was weighed daily and the amount consumed per day by each mouse was
calculated. This amount of food was then fed to a paired mouse in a second
untreated group (n=4). The food consumption of a control group of untreated
mice with free access to chow was also recorded. The body weights were
measured daily. The experiment was repeated with normal, age-matched C57
mice (n=5/group). Paired feeding studies were also performed with angiostatin
17
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
and endostatin treated mice following this protocol.
Cycled Treatnient ivith TNR-470 Eight week old, ob/ob mice (n= 15) were
treated with intermittent cycles of 10 mg/kg/day of TNP-470. Mice started at a
weight of 45 gm and were treated until they reduced to the body weight of age-
matched C57 mice. Treatment was then discontinued and the mice were
permitted to gain back to the starting weight of 45 gm. This cycle was
repeated
twice. Having demonstrated reproducibility here and efficacy in ob/ob mice
weighing greater than 60 gm, the treated group was then permitted to gain to
an
average weight of 59 gm on the third cycle (day 135) before restarting
treatment.
Subsequently, the mice were cycled between the weights of the ob/ob and C57
control groups. All mice were weighed daily.
Angiostatin and Enrlostatin Treatnlent of'Ohese Mice Recombinant murine
endostatin and angiostatin were produced by Escherichia coli as described by
O'Reilly et al, 1994 Cell. 79:315-328; O'Reilly, et al.. 1997. Cell. 88:277-
285.
Eight week old, ob/ob mice (n=3/group) received subcutaneous injections of
antiostatin at 20 mg/kg/day or 50 mg/kg twice a day, or endostatin at 50 mg/kg
twice a day for 22 days. A control group of ob/ob mice (n=3) received 0.1
ml/day of saline.
Body ConiPosition Assay The body composition of the mice was assessed using
a tritiated water dilution technique (Logsdon, 1972 Radiologic Technology. 44:
146-149). Five month old, treated (n=6) and control (n=3) animals from the
long-term treatment study with TNP-470 (day 90) and normal age-matched C57
mice (n=3) were used for this study. The mice were fasted for 12 hours with
free access to water. They were then gavaged with 200 l of tritiated water
(100
gCi/ml) and remained fasting for an additional 2 hours to allow the ti-itium
to
equilibrate within the water compartments. Animals were then anesthetized and
placed under a heating lamp for 2 min. to induce tail vasodilation. A 0.5 mm
deep incision was made perpendicular to the ventral long axis of the tail, 5
mm
from the base using a #10 surgical blade. Care was taken to avoid injuring the
central tail artery. A 50 1 blood sample was collected from the tail veins
into
serum separator vials and placed on ice. Bleeding was stopped by gentle
1O'
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
compression over the incision. Tritiated water specific activity was measured
in
samples containing 10 l plasma and 3 ml of Ecolume scintillation fluid (ICN;
Costa Mesa, CA) using a Wallac 1409 liquid scintillation counter. Variability
between duplicate samples averaged less than 10%. 'The followin~ t~~r-inulas
were used to calculate the desired parameters.
(1) Total Body Water = (standard activity) x 2,000
(sample activity - background activity) x 1.064
The factor of 1.064 is used to correct for the fact that serum is about 94%
water.
The factor of 2,000 is the dilution factor needed to calculate the total
tritium
activity in the dose administered.
(2) Lean Body Mass = Total BodxWater
0.73
The factor of 0.73 is the mean percent of total body water found in lean body
mass as derived from analysis of cadavers (Widdowson, et. al. 1968).
(3) Fat Mass = Total Body Weight - Lean Body Mass
(4) Percent Body Fat = Fat Mass x 100
Total Body Weight
Serunz Glucose Measuren:ents Serum glucose levels of the 9 month old,
treated (n=5) and untreated (n=13) ob/ob mice were measured at the end of the
weight response study. Blood samples (100 l) were collected from the tail
veins of anesthetized mice as described above. The serum glucose levels were
measured by the clinical chemistry laboratory at Children's Hospital, Boston,
and are expressed as mg/dl.
Cell Culture Studies
Bovine capillary endothelial cells (BCE) were obtained as previously
19
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
described by Folkman et al, 1979 Proc Natl Acad Sci. USA 76:5217-5221. BCE
were plated onto tissue culture plates coated with 1.5% gelatin and cultured
in
DMEM containing 10% bovine calf serum (BCS), 1% glutamine/antibiotics
(GPS), and 3 ng/ml recombinant human basic fibroblast growth factoi- (bFGF;
Scios Nova, Mountainview, CA). BCE were maintained at 37 C in 10% CO2
and were passaged at confluence using a 0.5% trypsin solution. 3T3-L! cells
were obtained from ATCC and maintained according to the accompanying
instructions. 3T3-L1 were cultured in DMEM containing 10% BCS without
antibiotics and plated onto uncoated tissue culture plates. Cells were
passaged
when 70% confluent to avoid differentiation in adipocytes using a 0.5% trypsin
solution.
Bovine Capillary Endotltelial Cell Prolif'erotion Assay BCE were maintained
at confluence for 5 days prior to use in the proliferation assays. The cells
were
washed with PBS and dispersed in a 0.05% trypsin solution. BCE were
resuspended in DMEM containing 10% BCS and 1% GPS (25,000 cells/ml) and
plated onto gelatinized 24-well culture plates at a density of 12,500
cells/well
(0.5 ml/well). The plates were incubated for 24 hr. The media was replaced
with 0.25 ml of assay media (DMEM + 5% BCS + 1% GPS) containing the test
sample at 2x the final concentration. After 30 min. of incubation, 0.25 ml of
assay media containing 2 ng/ml of bFGF was added. The wells then contained
0.5 ml of assay media with the test sample at Ix concentration and I ng/ml of
bFGF. After 72 hr of incubation, cells were dispersed in trypsin, resuspended
in
Hematall (Fisher Scientific, Pittsburg, PA) and counted using a Coulter
Counter.
3T3-L1 Cell Prolif'eration Assay Subconfluent cultures of 3T3-L I cells were
washed with PBS and dispersed in a 0.05% trypsin solution. The cells were
resuspended in DMEM containing 10% BCS (16,000 cells/ml) and plated onto
24-well culture plates at a density of 8,000 cells/well (0.5 ml/well). The
plates
were incubated for 24 hr. The media was replaced with 0.25 ml of assay media
(DMEM + 5% BCS) and the test sample at 2x the final concentration. After 30
min. of incubation, 0.25 ml of assay media containing 2 ng/ml of bFGF was
added. The wells then contained 0.5 ml of assay media with the test sample at
CA 02307792 2005-06-08
WO 99/39702 PCT/US98/22998
1 x concentration and I ng/inl of bFGF. After 72 hrs of incubation, cells were
dispersed in trypsin, resuspended in HematallT"" and countered using a Coulter
CounterTM
Immunohistochemistiy
A double fluorescent labeling technique was used to enable simultaneous
identification of proliferating cells (Brdu) and endothelial cells (von
Willebrand
factor; vWF) in adipose tissue sections. Adipose tissue was fixed for 4 hours
in
cold Carnoy's fixative and transferred to 100% ethanol. Tissues were embedded
in paraffin following standard histological procedures. Sections (5 m thick)
were permeabilized with 10 pg/ml proteinase K at 37 C for 20n min. and
washed with PBS. To label the proliferating cells, sections were iilcubated
with
anti-Brdu/nuclease mixture (Brdu Labeling and Detection Kit l; Boehringer
Mannheim, Germany) at room temperature for 60 min. and then washed with
PBS. Slides were incubated in fluoresceinated horse anti-mouse IgG (Vector;
Burlingane, CA) at 4 C overnight. To label the endothelium, sections were
blocked with 5% horse serum for 30 min. and then incubated with rabbit
antiserum against human von Willebrand factor (DAKO) diluted in 5% goat
serum in PBS. Antibody binding was detected by Texas Red-conjujated goat
anti-rabbit IgG (Vector Laboratories). Stained sections were preserved with
Fluoromount G (Southern Biotechnology Associates, Inc.). Slides were
analyzed using a Zeiss Axiophot fluorescent microscope (Zeiss, Oberkoclien,
FRG) with blue light at 485 nm for FITC excitation (Brdu) and green light at
:510-560 nm for Texas Red excitation (endothelium). Photographs were taken at
40X magnification with Ecktachrome p1600 film (Eastman Kodak_ Rochester.
NY). Proliferatingendothelium wei-e detectable on the film as yellow cells
(green + red).
Statistical Analysis
A one-tailed Student's t-test was used for comparison of the body
weights, percent body fat, and serum glucose levels between the treated and
untreated groups. Comparison between the treated and paired fed groups was
rnade using a paired Student's t-test.
2 i
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
Results
Weight Gain in Obese and Non-obese Mice
Both obese (ob/ob) and non-obese (C57) mice used in these siudies
experienced a period of rapid growth and weight gain during the first few
months of life, which slowed as the animals aged. However, the ob/ob mice
gained 2-3 fold more weight at much faster rates compared to the age-matched
controls. This difference was largely accounted for by the accumulation of
excess adipose tissue. In these studies, 8 week old ob/ob mice weighed 44.6 ~
0.7 gm and gained at an average rate of 0.4 gm/day (n+12). For comparison,
normal, age-matched C57 mice weighed 23.0 0.6 gm and gained at an average
rate of 0.1 gm/day (n=5). Obese mice reached maximum weights of '~reater than
70 gm at 8 - 9 months of age, compared to their normal counterparts which
reduced maximum weights of greater than 30 gm at 3- 4 months of age.
Body Weight Responses of Obese Mice to Angiogenesis Inhibitors
The effect of angiogenesis inhibitors on the ability of ob/ob mice to gain
weight was tested. TNP-470, a selective angiogenesis inhibitor, was
administered to obese mice at doses of 2.5, 5.0, 7.5, and 10 mg/kg/day for 21
days. Obese mice receiving TNP-470 lost weight at a rate, for a duration, and
to
an extent that were dose dependent (Figure 2). Rates of weight loss ranged
from
0.5 gm/day at a dose of 2.5 mg/kg/day to a maximum of 1.2 gm/day at doses
equal to or greater than 10 mg/kg/day (p <0.00001). For comparison, the
maximum rate of weight loss occurring in mice with no caloric intake for 4
days
was 1.2 g/day. In spite of continued treatment, the mice leveled or nadired at
body weights ranging from 40.5 0.9 gm after 10 days on 2.5 mg/kg/day
(p<0.000001). Relative to the average starting weight of 45 gm, this amounted
to weight losses of 4 - 25 gm, depending on the dose of TNP-470. ln contrast
the untreated mice gained an average of 9 gm, reaching a weight of 53.8 0.6
gm by the conclusion of this study. Relative to the final weights of the
control
mice, the TNP-470 treated mice weighed 13-34 gm less.
The body weights of obese mice were also evaluated in response to
specific angiogenesis inhibitors. Mice were treated with angiostatin at doses
of
22
CA 02307792 2000-04-27
WO 99/39702 PCTIUS98/22998
20 mg/kg/day or 50 mg/kg twice a day, or endostatin at a dose of 50 mg/kg/day
for 21 days (Figure 3). Control mice gained 0.4 gm/day from an average
starting weight of 45 gm, reaching 54.6 0.2 gm by day 22. tn eontrast, the
mice receiving angiogenesis inhibitors gained relatively little or lost
weight.
The responses to angiostatin were dose dependent. At 20 mg/kg/day, weight
gain was restricted to one third that of controls and the mice reached 48.5
1.1
gm by the end of the study (p<0.03). At 50 mg/kg of angiostatin twice a day,
the mice reduced to 41.7 0.2 gm and plateaued (p<0.0004). Mice treated with
endostatin initially lost about 2 gm, then gradually gained to 48.9 0.8 gm
by
day 22 (p<0.002). Relative to the final control weights, treated mice weighed
5
- 13 gm less.
Long-ternt Treatment ivith TNR470
Long-term studies were performed to determine the pattern of body
weight response to chronic therapy. The mice were treated with 10 mg/kg/day
of TNP-470 for a total of 138 days and the body weights were measured daily
(Figure 4A). Initially, the animals lost weight at an average rate of 1.2
gm/day,
reducing to a nadir of 20.2 0.5 gm at day 20. This was followed bv a
liinited
period of rapid weight gain. The body weights leveled at approximatelN 32 ~,m
by day 54 of the study, and then paralleled the growth curve of nornial, aged-
matched C57 mice.
Animals were fasted for periods totaling 24 hours on days 89-90 of the
study to conduct tritiated water dilution experiments. This resulted in a
weight
loss in all of the groups. However, the treated mice did not regain this lost
weight, in spite of returning to an ad librium diet. This is reflected by a
dip in
the curve followed by re-establishment of the slope at a lower body weight.
At the conclusion of the study, treated mice weighed 34.8 1,1 gm,
compared with control ob/ob mice which weighed 71.0 0.9 gm (p<0.00001)
and normal, age-matched C57 mice which weighed 33.1 0.7 gm. Treated mice
weighed 36 gm less than untreated ob/ob mice.
Body Weight Responses of Obese Mice to TNP-470 at Five Months of Age
As obese mice age, they remain 2 - 3 times as heavy as nornial C'57 mice
23
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
and the rate of weight gain decreases, becoming more analogous to chronic
obesity in humans (Chlouverakis and Hojnicki, 1974). To determine whether
TNP-470 would be similarly effective under these conditions, ob/ob mice were
housed until they completed their rapid growth phase. The mice were 5 months
of age and weighed greater than 60 gm at the start of the experiment.
Untreated
mice gained weight at a rate of 0.1 gm/day reaching 73.0 1.0 gm by the end
of
the study. In contrast, the treated mice reduced at a rate of 0.2 gm/day to
39.0 f
1.9 gm by day 40 of treatment and leveled for the duration of the 67 day study
(p<0.00001) (Fig. 4). Normal C57 mice weighed approximatelv 34 arn at this
age. Therefore, these older, very obese, and more weight stable animals were
equally responsive to TNP-470.
Body Conzposition of Obese Mice Treated ivith TNP-970
As the only growing adult organ and site of angiogenesis in normal
animals, it was hypothesized that adipose tissue would be selectively
susceptible
to antiangiogenic therapy. The observation that the body weight of treated
animals leveled at the approximate weight of normal, age-matched C57 mice,
regardless of starting weight or duration of therapy was consistent with this
hypothesis (Figure 4A and 4B). A tritiated water dilution technique was used
to
assess the body composition of the mice and determine if the weight loss was
associated with a decrease in percent body fat (Logsdon, 1972). Mice treated
with TNP-740 (10 mg/kg/day) for 90 days (Figure 4A) and age-matched ob/ob
and normal, C57B1/6 mice were used. Weight loss in response to
antiangiogenic therapy was associated with a 46% reduction in percent body fat
from 35% to 19% (Table 2; p<0.001).
Table 2: Percent Body Fat
Group Mice/ rou Body ivei ht nz Body at %
Untreated ob/ob 6 67.4 0.8 3 5.2f2.2
TNP-470 ob/ob 6 3 0.6f0.7 * 18. 8t 1.3 **
Untreated C57BL/6 4 28.3 0.7 9.2 2.3
Treated (TNP-470; 10 mg/kg/day x 90 days) and control ob/ob mice, and age-
matched normal C57 mice were fasted and gavaged with a known quantity of
24
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
tritiated water. The radiolabel was allowed to equilibrate throughout the
water
compartments of the animal. The amount of label was then measured in a small
volume of plasma and the dilution factor was used to determine the total body
water, from which the percent body fat was calculated.
* p<0. 00001 ** p<0.001
Serum Glucose Lenels of Obese Mice Treated ivith TNP-470
The morbid obesity in ob/ob mice is associated with the development of insulin
resistance and hyperglycemia (Coleman and Hummel, 1973). To determine if
the weight loss resulting from antiangiogenic therapy protected against the
onset
of diabetes, serum glucose levels were measured in ob/ob mice treated with
TNP-470 for 115 days and the untreated controls. The untreated mice (73 gm)
had an average glucose level of 224 16 mg/dl. ln contrast, ob/ob mice
treated
with TNP-470 (41 gm) had an average glucose level of 158 12.6 mg/dl
(p<0.05). This is within the normal range of 62 - 175 mg/dl reported for this
species (Harkness and Wagner, 1995 The biology and medicine of rabbits and
rodents. Lea & Febiger Books. Media, PA. p. 93).
Table 2: Serum Glucose Levels
Group Mice/ rou Body ivei ltt nt I3orly at %
Untreated ob/ob 11 73.0 1.0 224 16
TNP-470 ob/ob 6 41.2t1.1* 158f13**
Blood samples were drawn from the tail veins of treated (TNP-470; 10
mg/kg/day x 115 days) and control ob/ob mice. Serum glucose levels were
measured in serum samples by the Clinical Chemistry Laboratories at Children's
Hospital, Boston. *p<0.00001 ** p<0.05
Cycled TNP-470 Treatment of Obese Mice
To demonstrate reversibility and reproducibility of antiangiogenic
therapy in this model of normal tissue growth, ob/ob mice were treated with
intermittent cycles of TNP-470 at a dose of 10 mg/kg/day (Figure 4C). Mice
started at a weight of 45 gm and were treated until they reduced to the body
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
weight of age-matched C57 mice. Treatment was then discontinued and the
mice were permitted to gain back to the starting weight of 45 gm This cvcle
was repeated twice. Having demonstrated reproducibility here and efficacy in
ob/ob mice weighing greater than 60 gm (Figure 4B), the treated group was then
permitted to gain to approximately 60 gm on the third cycle (day 135) before
restarting treatment. This is an ongoing study and the mice are being cycled
between the weights of the ob/ob and C57 control groups. The rate of weight
gain off therapy was greater than that of controls. Accelerated weight gain
has
been previously reported in these animals following caloric restriction
studies
until the normal obese weight is established (Chlouverakis, 1970). The treated
mice have similarly and repeatedly reduced while receiving TNP-470 and
regained when the drug was discontinued.
Conzparison of TNP-470 to Fenflurantine
The weight loss response to TNP-470 was compared to that of
fenfluramine, a serotonin uptake inhibitory with anorectic effects (Figure 5).
Obese mice treated with a high therapeutic dose (40 mg/kg/day)
ot'teiitluramine
lost 4 - 5 gm in the first 5 days of therapy (p<0.005). Thereafter, they
resumed
gaining weight gradually despite continued treatment. On day 45 of the study
the fenfluramine treated mice were divided into 3 groups and were either 1)
taken off therapy, 2) continued on fenfluramine, or 3) changed to 10 mg/kg/day
of TNP-470. By day 94, the control mice weighed 65.8 1.4 gm; mice taken
off fenfluramine weighed 65.6 + 1.4 gm; and mice continued on fenfluramine
weighed 62.4 + 1.9 gm. There was no significant difference between any of
these groups. In contrast, substituting TNP-470 for fenfluramine in a subgroup
of mice resulted in reduction to the weight of normal C57 mice, 38.5 0.7 gm
(p<0.000001). This is further illustrated by the appearance of the inice.
Untreated ob/ob mice are not easily distinguishable from those treated with
fenfluramine, whereas, mice receiving TNP-470 are the same size as their C57
counterparts.
Tolerance of Long-tern: Treatnzent with TNF-470 in Obese Mice
Treatment with TNP-470 has been very well tolerated in doses up to 10
26
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
mg/kg/day for periods up to 138 days, thus far. This is approximately 1/5 of
the
life span of the animals. The only adverse effect noted is mild, superficial
scaring at the site of repeated injections of the drug. Appetite is moderately
suppressed for the first 10 - 20 days of treatment and then gradually
normalizes.
The activity level of the treated mice was similar to that of C57 mice and in
contrast to control ob/ob mice which became inactive as they gained weight.
In conclusion, these studies have demonstrated that antiangiogenic
therapy results in significant weight loss and reduction in body fat in
genetically
obese mice using three distinct angiogenesis inhibitors. This was further
substantiated by the demonstration that inactive preparations of angiostatin
and
endostatin did not effect body weight. The following series of experiments was
directed at determining the mechanism(s). Effects on appetite and on the
peripheral tissues are being investigated. The results are summarized below.
Paired Feeding Studies ivith Obese Mice Treated ipith Angiogenesis Inhibitors
The effect of appetite on the body weight of TNP-470 treate/d mice was
isolated in paired feeding experiments. The precise amount of chow consumed
daily by each treated mouse was fed to a paired mouse in an untreated group
(Figure 6A and 6B). The body weights of the paired mice were used to isolate
and measure the effects of appetite changes in the treated animals. Obese
inice
receiving TNP-470 ate less for approximately the first 18 days of treatment.
This accounted for 56% of the initial weight loss. Thereafter, the appetite
gradually normalized in spite of continued treatment (up to 138 days), as
reflected by a steady gain in the paired mice to near control weights. To
determine if the anorexic effect was reversible and reproducible in the same
animals, a group of mice was treated with intermittent cycles of TNP-470 and a
paired-fed group was used to follow the impact of appetite changes (Figure
6A).
Mice received TNP-470 for 20 days, came off therapy for 15 days, and were
then treated a second time for 36 days. The anorexic effect was only present
during the initial exposure to TNP-470. During the second cycle of therapy ,
the
treated animals lost weight but maintained their appetite as reflected by the
steady increase in the weights of the pair-fed mice. Appetite accounted for
only
27
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
17% of the weight difference between the treated and untreated mice during the
second cycle. The anorexic mechanism is not yet determined.
There was no anorexic effect associated with endostatin or angiostatin at
(20 mg/kg/day). Angiostatin at 50 mg/kg twice a day was associated with a
small decrease in appetite accounting for less than 20% of the total weight
lost.
Paired Feerling Studies ivith Norntul C5 7 Mice Treated ivith TNR470
Normal C57 mice treated with TNP-470 also lost weight (Figure 6B). In
this case, nearly all of the initial weight loss was attributed to a decrease
in
appetite, as reflected by the paired-fed mice. The anorexic effect lasted
about 8
days, after which the appetite of the treated mice gradually increased and the
paired mice gained weight. The treated mice reduced to an average of 8S% the
control weight (p<0.00001). This is significantly less than that seen in the
ob/ob
mice on the same dose of TNP-470 (56% of control weight). This may reflect
the fact that these normal animals have a much lower percent of body fat,
which
appears to be the selective target of this therapy.
The weight loss in the treated animals which not be explained by
appetite, is most likely the result of a peripheral effect at the tissue
level. This
conclusion is further supported by the observation that weight gain of both
the
ob/ob and C57 mice was restricted by the angiogenesis inhibitors. The treated
mice gained relatively little, in spite of an increased food consumption
sufficient
to cause weight gain in the paired-fed groups.
Endothelial Cell Proliferation in Adipose Tissue of Obese and Normal Mice
To determine whether adipose tissue growth was associated with
endothelial cell proliferation, subcutaneous adipose tissue sections from
treated
and untreated ob/ob mice and normal C57 mice were examined. A double
fluorescent staining procedure was used to simultaneously identify both the
endothelium and any proliferating cells. The endothelial cell marker, von
Willebrand factor (vWF), was stained with a red fluorescent tag and Brdu in
proliferating cells was stained with a green fluorescent tag. Proliferating
endothelial cells were stained with both tags and appeared yellow.
The adipose tissue of normal C57 mice contained relatively small
28
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
adipocytes and a high density of endothelial cells as demonstrated by the
abundance of red stained cells. Proliferating endothelial cells were also
occasionally seen (yellow cells). The adipose tissue of untreated obese mice
contained numerous proliferating endothelial cells. The density of endothelial
cells was diluted by the large size of the intervening adipocytes in the obese
mice. The adipose tissue of treated ob/ob mice contained dilated, tortuous
microvessels with a lower percentage of endothelial cells and rare
proliferating
cells, consistent with inhibition of angiogenesis. Tissue sections from obese
mice taken off TNP-470 for one week were also examined. These aninials were
gaining weight at rates greater than control. The endothelial cell density is
notably less than that of normal adipose tissue. However, there is an
abundance
of proliferating cells. Proliferating endothelial cells (yellow) as well as
non-
endothelial cells (green) are present. Dilated microvessels, similar to those
seen
in the treated animals, were also observed.
Endotl:elircl Cell and 3T3-L1 Cells Proliferation In Response to TNR-470
To determine whether TNP-470 may be having a direct effect on
preadipocytes, proliferation assays using 3T3-Ll cells were performed. This
cell line has most of the attributes of adipose cells in vivo (Green 1978. The
adipose conversion of 3T3 cells. In: 10th Miami Symposium on differentiation
and development. Ahmad F, Schultz J, Russell T, Werner R (eds). Academic
Press, New York. p. 13.). Cells were incubated in a range of TNP-470
concentrations for 72 hours and subsequently counted. Comparisons were made
to bovine capillary endothelial cells (BCE). The half-maximal cytostatic
inhibition of endothelial cell proliferation was at approximately 100 pg/ml.
In
contrast, 10 g/ml of TNP-470 was required for a similar degree of inliibition
in
the 3T3-L1 cultures.
Conclusions
These studies demonstrate that an adipose tissue growth and mass is
angiogenesis dependent; angiogenesis inhibitors reduce body weight and
adipose tissue mass; reduction of body weight with antiangiogenic agents is
dose dependent, reversible, and occurs regardless of starting weight; weight
29
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
reduction was achieved with every angiogenesis inhibitor tested; TNP-470
results in a transient decrease of appetite; long-term treatment with
angiogenesis
inhibitors is well tolerated; treatment of morbid obesity with angiogenesis
inhibitors prevents hyperglycemia.
Example 3: Treatment of Animals withCardiac Hypertrophy using
Angiogenic Inhibitors.
Model
Chronically elevated thyroid hormone levels result in volume overload
of the heart leading to eccentric cardiac hypertrophy in both animals and man
(Osler W. 1892. The principles and practice of medicine. D. Appelton and
Company. New York, pp. 712-714). Myocardial hypertrophy is usually
associated with abnormalities of the coronary vascular system and eventual
loss
of ventricular function (Tomanek and Hovanec, 1981 J Mol Cell Cardiol.
13:471-488; Friberg, 1988 Cardiovasc. Res. 22:329-339.). However, thyroid
hormone induced cardiac hypertrophy is unique in that there is frequently
enhanced ventricular function and a proliferation of the supporting
microvascular system (Tomanek et al, 1998 Circ Res. 82:587-593).
It was hypothesize that the growth of cardiac tissue in this model of
myocardial
hypertrophy may be mediated by the vascular endothelium and could be
suppressed with the use of angiogenesis inhibitors. The implications of such a
finding relate to the role of the vascular endothelium as a regulator of
normal
cardiac mass and function.
Methods
Animal studies were carried out at Children's Hospital, Boston in accordance
with institutional guidelines. Male C57BL/6J mice were purchased at 6 weeks
of age and acclimated for one week before experimentation (Jackson
Laboratories, Bar Harbor, ME). Mice were caged collectively and had free
access to water and standard chow. Animals were anesthetized with inhaled
methoxyfluorane prior to procedures causing discomfort or distress and were
monitored until fully awake. The hair was shaved to facilitate injections.
Mice were treated with thyronine (0.2 mg/kg/day, i.p.) for 12 days to induce
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
cardiac hypertrophy. In addition, mice simultaneously received s.c injections
of
saline or TNP-470 at either 10 mg/kg or 20 mg/kg every other day (n=3/group).
Results were compared to controls treated with saline alone (no thvronine).
Animals were weighed at the end of the study and sacrificed by cervical
dislocation. The hearts were immediately removed and placed in warmed buffer
solution while still beating to empty the organ of remaining blood. The
ventricles were dissected clean of the great vessels, atria, and connective
tissue
and the chambers were opened. The tissues were blotted dry and wei-ghed.
Ventricular weights were normalized to body weight and presented as percent of
body weight.
Results
Thyronine treatment resulted in cardiac hypertrophy, increasing the
ventricular
weight from 0.38% to 0.58% of body weight. This effect was blunted by TNP-
470 in a dose dependent fashion (Figure 7). At a TNP-470 dose of 20 mg/kg,
the ventricular weight achieved with thyronine treatment was significantly
less
than controls, at 0.51% of body weight (P<0.05). Furthermore, these figures
may underestimate the effects of TNP-470, given that there was a weight loss
of
1-2 gm in the mice receiving TNP-470 during the 12 day study. The decrease in
weight results from a decrease in adipose tissue mass. However, the cardiac
size
of the treated animals is expected to be appropriate for their higher starting
weight. Studies to verify this are in progress. The doses tested thus far are
2/3
less than typically used to treat cancer in mice. A greater reduction in
hypertrophy is expected with higher doses.
Example 4: Treatment of Intestinal Polyposis with Angiogenesis
Inhibitors.
Model
Ap?'" mice serve as a model of intestinal adenomatous polyps. Min (multiple
intestinal neoplasia) is an ethylnitrosourea (ENU)-induced mutation in the
murineApc (adenomatous polyposis coli) gene (Moser et al, 1990). It is similar
to the germline mutations in the APC' gene in humans with familial adenomatous
polyposis or Gardner Syndrome (Joslyn et al, 1991 Cell. 66:601-6 ] 3). These
are
31
CA 02307792 2000-04-27
WO 99/39702 PCTIUS98/22998
inherited conditions characterized by intestinal polyps that eventually
transform
into cancer. APC is the most frequently mutated gene in sporadic human colon
cancer, as well (Nishisho et al, 1991 Science. 253:665-668).
Heterozygous C57B 1/6J Min/+ mice spontaneously develop 30-60
adenomas throughout the intestine during the course of their lifetimes (Moser
et
al, 1990 Science. 247:322-324). The polyps continue to grow in size,
ultimately
leading to the death of the animals by about 120 days of age.
It was hypothesized that the growth of these pre-neoplastic lesions is
dependent on the vascular endothelium and could be suppressed witli the use of
angiogenesis inhibitors.
Methods
Animal studies were carried out at Children's Hospital, Boston in
accordance with institutional guidelines. Male Min/+ (C57BL/6J) mice were
purchased at 6 weeks of age and acclimated for one week before
experimentation (Jackson Laboratories, Bar Harbor, M:E). Mice were caged
collectively and had free access to water and standard chow. Animals were
anesthetized with inhaled methoxyfluorane (Pitman-Moore, Inc., Mundelein, IL)
prior to procedures causing discomfort or distress and were monitored until
fully
awake. The hair was shaved to facilitate injection.
TNP-470 was stored dry at 40 C. Solutions were prepared fresh in
normal saline and administered subcutaneously in rotating sites along the
dorsum. Mice were treated with 10 mg/kg of TNP-470 every other dar tur I or
3 weeks (n=3/group). A control group received 0.1 ml of saline everv other day
(n=3).
Chow was withheld for 12 hours prior to the termination of each study to
clear the intestinal tract. Animals were sacrificed by continuous inhalation
of
CO2. The entire intestinal tract was removed and flushed with phosphate
buffered saline. The intestine was opened longitudinally and the luminal
surface
was examined for the number, size, and location of polyps using a dissecting
microscope (magnification 18 X). Tissue samples were fixed in buffered
formalyn and embedded in paraffin. Sections were stained with H&E for
32
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
histologic evaluation.
Results
The number and size of intestinal polyps was significantly reduced in
Min/+ mice treated with an angiogenesis inhibitor. Following 3 weeks of
treatment, control mice had 23 t 1.5 polyps, while mice treated with TNP-470
had 8 1.7 polyps (Figure 8A). This is a reduction of 66% (p<0.05). The
diameter of the polyps in the control mice averaged 1.90 0.86 mm (range
0.075 to 0.425 mm) (Figure 8B). In contrast, the diameter of the polyps in the
treated mice averaged 1.02 0.40 mm (range 0.05 - 0.175 mm).
Histologic examination of polyps from the control mice demonstrated
marked cellular dysplasia and numerous cavernous vessels. Lesions were
polyploid, extending from the mucosa into the lumen and occasionally outward
through the visceral surface. In contrast, polyps from the treated mice
retained
morphologic similarity with the surrounding mucosa, with fewer dysplastic
cells
and fewer, smaller vessels.
Conclusion
The occurrence, growth, and possibly transformation of intestinal
adenomatous polyps is angiogenesis dependent and regulated by the vascular
endothelium.
Example 5: Treatment of Animals witli Endometriosis.Using Angiogenic
Inhibitors.
Model
Up to 50% of all menstruating women may be afflicted with
endometriosis (Williams and Pratt, 1977 Am J Obstet Gynecol. 129:245.).
Human endometriosis can be surgically mimicked in a rodent model (Cummings
and Metcalf, 1996 PSEBM. 212:332). Studies were designed to test whether the
growth of these tissue implants is dependent on the vascular endothelium and
could be suppressed with the use of angiogenesis inhibitors.
Methods
Animal studies were carried out at Children's Hospital, Boston, MA, in
accordance with institutional guidelines. Female (B6C3F1) mice were
33
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
purchased at 8 weeks of age and acclimated for one week before
experimentation (Jackson Laboratories, Bar Harbor, ME). Mice we--e caged
collectively and had free access to water and standard chow. Animals were
anesthetized with inhaled methoxyfluorane (Pitman-Moore, lnc., Mundelein, IL)
prior to procedures causing discomfort or distress and were monitored until
fully
awake. The hair was shaved to facilitate injections.
TNP-470 was stored dry at 40 C. Solutions were prepared fresh in
normal saline and administered subcutaneously in rotating sites along the
dorsum. Mice were treated with 10 mg/kg of TNP-470 every other day for 3
weeks (n=4/group). A control group received 0.1 ml of saline every other day
(n=3).
Using aseptic technique, mice were anesthetized and the abdominal hair
was shaved. A midventral incision was made and the left uterine horn was
exposed and ligated at both ends. The uterine horn was excised, placed in
warmed media (Ham's F12) and opened longitudinally. Pieces of tissue were
cut 2 mm in diameter using a dermal punch biopsy. The tissue pieces were
position on the intestinal mesentary (4 pieces/animal). Implants were held in
place with a fibrin glue preparation, The abdominal incision was closed and
the
mice were permitted to recover for 48 hours. Thereafter, the mice were divided
into two groups of 4 mice each. The control group was treated with saline. The
treatment group received TNP-470 (10 mg/kg, every other day). The animals
were sacrificed by continuous inhalation of CO2 three weeks following surgery.
The peritoneal cavity was examined for endometrial tissue. The tissue was
counted and the diameters were measured in the x and y dimensions using
calipers and averaged.
Results
The size of the endometrial tissue was reduced in mice treated with an
angiogenesis inhibitor. Following 3 weeks of treatment, control mice
endometrial tissue measured 2.9 0.4 mm, while mice treated with TNP-470
endometrial tissue measured 1.9 0.5 mm, as shown by Figure 9. The number
of tissue implants was similar between the two groups.
34
CA 02307792 2000-04-27
WO 99/39702 PCT/US98/22998
Conclusion
The data indicates that the growth of endometrial tissue implanted into
the abdominal cavity is suppressed by angiogenesis inhibitors. The dose of
TNP-470 used in this pilot study is 1/3 of the established tumor dose.
